Vehicle Counting And Traffic Congestion Detection Using YOLOv3

Abstract

The purpose of this project is to utilize Python modules and a sophisticated deep learning algorithm to identify, categorize, track, and tally moving vehicles from highway CCTV footage. Additionally, the system predicts traffic congestion by analyzing the number of vehicles in consecutive video frames. When congestion is detected, the application automatically notifies the traffic police who receive a message on their mobile, prompting them to address the existing traffic jam in the area. The project's core involves a vision-based vehicle detection and counting system that relies on the YOLOv3 model and openCv-python library to achieve its objectives.